google-adk

MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).

• Third-party content exposure detected (medium risk: 0.30). In the FastAPI runtime path, POST /chat and GET /stream pass outsider-authored free text (request.message or query param message) directly into runner.run_async(... new_message=types.UserContent(parts=[types.Part(text=request.message)])), which the model ingests without any intermediary selection step.